In 1989, I took my first decision sciences course, and started coding in SAS at the age of 20. I greatly enjoyed pulling discoveries buried within mounds of data, although and even small datasets had many discoveries back then.
At the root of every model I’ve built, even the simplest, was a solid understanding and foundational rigor of statistical theory. When computing simple statistics or developing descriptive models, I thought through the math behind the model and how this would impact the formation, application, and interpretation.
It was about 30 years ago when I started my decision sciences journey, and I’m still applying techniques and building models to empirically solve problems, answer questions, overcome challenges that improve, reduce error, or otherwise benefit a situation.
Over the past three decades, I’ve noticed trends and shifts, an evolution of sorts, in the foundational underpinnings of development and application within this interesting profession. I’ve come to the following conclusions that illustrate the evolution of the data science function over the past few decades: